© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 15469
    Prev
    Next
    INDEX
    Explanations

    references to dissidents or perceived opponents within a larger conflict

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ocupa
    -0.30
     arché
    -0.28
     déc
    -0.27
     rising
    -0.27
     ingenieros
    -0.27
     descans
    -0.26
     ocup
    -0.26
    features
    -0.25
    anlage
    -0.25
     собе
    -0.25
    POSITIVE LOGITS
    RegressionTest
    0.93
     <<<<<<<<<<<<<<
    0.72
    fjspx
    0.72
     nakalista
    0.72
    DataAnnotations
    0.69
    jsxFileName
    0.69
    SaveChangesAsync
    0.68
    addContainerGap
    0.67
    UnsafeEnabled
    0.67
     accusations
    0.66
    Activations Density 0.465%

    No Known Activations